As is known in the art, power electronics is a key technology for addressing the energy challenges. Improvements in performance of power electronics coupled with their expanded use could lead to dramatic reductions in electricity consumption (as much as 20-30% by some estimates). However, achieving such reductions in electricity consumption requires systems having increased efficiency while at the same time being smaller and less expensive than existing systems. Such power electronics are important both for reducing consumption of energy (though the improved capability and efficiency of loads and sources) and for improving the efficiency of an electrical grid itself (e.g., though improvements in power factor). Presently, power electronics at a grid-interface level efficiencies typically in the range of about 70-90% at full load. Furthermore, this efficiency typically falls off rapidly at reduced loads, such that average efficiency, as well as losses, are lower than the 70%-90% range. It has been estimated, for example, that power supply losses account for 20 to 70% of all energy that electronic products consume. Likewise, poor power factor is estimated to be indirectly responsible for as much as 2.8% of energy consumption in commercial buildings. There is thus a need to provide power electronics having improved peak and average efficiencies, an improved power factor and also having reduced size, weight and cost to enable greater adoption and utilization.
One area in which such improvements can be made is in power supplies that provide the interface between high-voltage DC or AC inputs (i.e., grid-scale voltages) and low-voltage DC outputs. This includes supply of energy from an AC grid to DC loads such as computers, electronic devices and LED lighting, which represents both a large use of electrical energy and a place where substantial energy is presently lost. It has been reported that over 28% of domestic electrical energy usage goes into “miscellaneous” loads—including electronic devices with power supply front ends. Furthermore, the percentage of energy going to electronic loads is growing at twice the rate of other loads. As much as 4% of the entire U.S. national energy consumption can be traced to power supply losses for electronic loads, owing largely to poor average power supply efficiencies. Moreover, the lack of power factor correction in most power supplies yields additional indirect (system-level) losses that can be quite substantial. This represents a tremendous waste of energy and the generation of unnecessary emissions.
Improved power conversion is also important for DC-input systems. DC distribution (nominally at 380 V DC) is sometimes considered an alternative to AC distribution in commercial buildings and data centers, as it offers higher efficiency, more effective management of power factor correction, and easier integration of distributed renewable sources and energy storage. For example, early demonstrations have shown that 380 V DC distribution architectures can result in energy savings of around 15% over standard 208 V AC distribution in data centers. This is due to the higher distribution voltage and fewer voltage conversion stages required in DC distribution systems.
The effectiveness of DC distribution depends, at least in part, upon DC-DC converters that convert the voltage from 380 V (actually 260 V-410 V) to the lower voltages needed for lighting and electronic loads. In data centers, where such DC distribution architectures are expected to be first deployed, the 380 V DC source will need to be converted to 12 V using a DC-DC converter located in each “rack” (i.e. a frame or enclosure for mounting multiple equipment module) to power the servers. To reduce cost in such an architecture, extremely high-efficiency converters (e.g., converter having efficiencies in the range of 97% and above) will be needed.
High-performance power converters that deliver energy from high-voltage DC or AC sources to low-voltage DC loads, with input voltages consistent With the AC grid (e.g., up to 240 V AC) and DC distribution systems (260-410 V DC) and output voltages of volts to tens of volts, achieve efficiencies of up to 90-95% (5-10% loss) with much lower average efficiencies. Technologies that achieve greater efficiency, power factor, and miniaturization of power converters that deliver energy from high-voltage DC or AC sources to low-voltage DC loads can be difficult to design, as the high input voltage and large voltage conversion ratios can result in large semiconductor switch and magnetic core losses (e.g., in transformers and inductors), and the wide operating ranges of input voltage (e.g., 1.6:1 or more) and power (e.g., 10:1 or more) place constraints on many design techniques.
One technique for achieving high efficiency power conversion within a switched converter includes using zero-voltage switching (“ZVS”), in which the transistor voltage is constrained close to zero when the transistor switches on or off, and/or zero-current switching (“ZCS”), in which the transistor current is constrained close to zero when the transistor switches on or off. Without soft switching, transistor switching loss can reduce the efficiency of the converter and limits power density.
Unfortunately, while available soft-switching circuits can achieve very high efficiencies under specific operating conditions, performance tends to degrade greatly when considering requirements of operation across widely varying input voltage and power levels. In particular, with conventional circuit designs and control methods, it is difficult to maintain desirable circuit waveforms (e.g., ZVS/ZCS switching, minimum conduction current, etc.) as power is reduced from maximum and as the input voltage varies from nominal.
For example, one technique for controlling resonant soft-switched inverters (e.g., series, parallel, series-parallel, LLC converters, etc.) is a frequency control technique, in which an output voltage is regulated in the face of load and input voltage variations by modulating a converter switching frequency. Because of the inductive loading requirements to achieve ZVS switching which is important at high voltage levels, power is reduced in such converters by increasing switching frequency, which can exacerbate switching loss. Wide frequency operation also makes design of magnetic components and EMI filters more challenging. Moreover, depending upon resonant tank design, circulating currents in the converter may not back off with power, reducing power transfer efficiency.
An alternative method that can be applied to bridge converters at fixed frequency is phase-shift control, in which the relative timing of multiple inverter legs are modulated to control power. However, conventional full-bridge resonant converters using phase shift control suffer from asymmetric current levels between the two inverter legs at the time of switching as the legs are outphased to reduce output power. The result is that the transistors in the leading inverter leg start to turn-off at large currents. Also, as outphasing is increased, the transistors in the lagging inverter leg lose ZVS turn-on capability. These factors result in extra losses and lead to lower converter efficiency at partial loads, and consequently to poor design tradeoffs.
Other fixed frequency control techniques, such as asymmetrical clamped mode control and asymmetrical pulse width control, have also been developed. However, these techniques also lose zero voltage switching (ZVS) capability as the output power is reduced. Hence, they also do not maintain high efficiency across a wide load range. There is an evident need for circuit designs and associated controls that can provide reduced loss when operating over wide input voltage and power ranges, and which can provide large step-down voltage conversion.
Frequency multiplier circuits can be used in extremely high-frequency RF applications and are sometimes used in switched-mode inverters and power amplifiers. Frequency multiplier circuits are not typically used in DC-DC converters, however, because the output power of a frequency multiplier inverter is inherently low relative to required device ratings.